malaria_detection.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "malaria-detection.ipynb",
      "provenance": [],
      "authorship_tag": "ABX9TyOiys8YYL9oUwcFubDaSyqX",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "f79e1dee556544cfb875121aef27c376": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_3e57e7591462419395966022a00f3492",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_363adaa5bf3643d8a925453c60e20b77",
              "IPY_MODEL_983fb7768bfc4af1ab2784cb97246931"
            ]
          }
        },
        "3e57e7591462419395966022a00f3492": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "363adaa5bf3643d8a925453c60e20b77": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_2ed1c4121841437f8317486bbd8d02f6",
            "_dom_classes": [],
            "description": "Dl Completed...: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_46e7c84750f146419a7b7e94d67263c8"
          }
        },
        "983fb7768bfc4af1ab2784cb97246931": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_3d6fbd4aeb4c4762a1ae335d366a5503",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1/1 [00:13&lt;00:00, 13.26s/ url]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_1b718f447be848248c8c93c9c69f6244"
          }
        },
        "2ed1c4121841437f8317486bbd8d02f6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "46e7c84750f146419a7b7e94d67263c8": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "3d6fbd4aeb4c4762a1ae335d366a5503": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "1b718f447be848248c8c93c9c69f6244": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "e1d2814c9b72448f987b71f558a2f4cb": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_534d52c5aeb844a7b86f4dadbeb5b14f",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_4e415db71c68415a9110e0721681e8d4",
              "IPY_MODEL_89f9120489cd4ac8b8c84a65c7b38822"
            ]
          }
        },
        "534d52c5aeb844a7b86f4dadbeb5b14f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "4e415db71c68415a9110e0721681e8d4": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_5e46c33c819041cfa50376e81254ac39",
            "_dom_classes": [],
            "description": "Dl Size...: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_50ab3fb0afeb4261983383f8b95529e6"
          }
        },
        "89f9120489cd4ac8b8c84a65c7b38822": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_d45b48af493840f4b767d09e97fb6c55",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 337/337 [00:13&lt;00:00, 25.49 MiB/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_bbfe2e6d491b4cf1864f1388a32b8aa1"
          }
        },
        "5e46c33c819041cfa50376e81254ac39": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "50ab3fb0afeb4261983383f8b95529e6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "d45b48af493840f4b767d09e97fb6c55": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "bbfe2e6d491b4cf1864f1388a32b8aa1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "ea80d6e88f6b493aa0feb35fcaa70124": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_4e7df5b9fafa4ac388da1be19dcae056",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_bcf4803447d84894b8deedfe64c85c65",
              "IPY_MODEL_353701ba92fa419b90a9c88d331cc807"
            ]
          }
        },
        "4e7df5b9fafa4ac388da1be19dcae056": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "bcf4803447d84894b8deedfe64c85c65": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_01583e8bf90d4fcabf4da636b586922e",
            "_dom_classes": [],
            "description": "Extraction completed...: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_cb600d99f4a942a694716c56e3c3e5a0"
          }
        },
        "353701ba92fa419b90a9c88d331cc807": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_06e89eb94cc64fe2aafc413a925bb854",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1/1 [00:13&lt;00:00, 13.18s/ file]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_bb17062811234f44b5e9ba875da45e77"
          }
        },
        "01583e8bf90d4fcabf4da636b586922e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "cb600d99f4a942a694716c56e3c3e5a0": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "06e89eb94cc64fe2aafc413a925bb854": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "bb17062811234f44b5e9ba875da45e77": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0cd1d3ba1dd74d2f98445758392db6ec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_e182dd7451b7464f99ea578854597525",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_645e22283c8c4d9081b901dba6185169",
              "IPY_MODEL_78e3b6f820e54bd29df7a1e3102b54bd"
            ]
          }
        },
        "e182dd7451b7464f99ea578854597525": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "645e22283c8c4d9081b901dba6185169": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_8d6f1f2d1e7743f7997f6c4a67bc5cb7",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "info",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_0b76d70cc7b84e2ebeeea210ae189ac5"
          }
        },
        "78e3b6f820e54bd29df7a1e3102b54bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_8558ea86176f4f77922a9b767b477a30",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 27558/0 [00:09&lt;00:00, 2731.40 examples/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_1c7963b3fddb4c6fac1de936488f395a"
          }
        },
        "8d6f1f2d1e7743f7997f6c4a67bc5cb7": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "0b76d70cc7b84e2ebeeea210ae189ac5": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "8558ea86176f4f77922a9b767b477a30": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "1c7963b3fddb4c6fac1de936488f395a": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "73a0ee47ab2a4f9f9fe160d901a3aa47": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_3393b45012be4a99aa69f00507d654c1",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_b3c1dc28d48342eeabcc10c847023ebd",
              "IPY_MODEL_cedc522d27f24be58fea368900bfe447"
            ]
          }
        },
        "3393b45012be4a99aa69f00507d654c1": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "b3c1dc28d48342eeabcc10c847023ebd": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_d37c1169e9634f98ad0bca7e3293633d",
            "_dom_classes": [],
            "description": " 89%",
            "_model_name": "FloatProgressModel",
            "bar_style": "danger",
            "max": 27558,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 24599,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_ec5c8f36f679419db79f28a32ea947de"
          }
        },
        "cedc522d27f24be58fea368900bfe447": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_e956e91be6e94fcaa6728b6e62158414",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 24599/27558 [00:00&lt;00:00, 28942.55 examples/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_8f102d253fe841a09af03abf90a63967"
          }
        },
        "d37c1169e9634f98ad0bca7e3293633d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "ec5c8f36f679419db79f28a32ea947de": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "e956e91be6e94fcaa6728b6e62158414": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "8f102d253fe841a09af03abf90a63967": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "8cc0c81d1f54464eb4668fe3e64aeeba": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_69e0bbf660c14691873bd2809f290be7",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_3e98141c909d4463835c530846502760",
              "IPY_MODEL_2dc78bf5fe0a4c02b75af9e5b2a6a0dc"
            ]
          }
        },
        "69e0bbf660c14691873bd2809f290be7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "3e98141c909d4463835c530846502760": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_d8070f6d74844ae197c0f8ef6a82c053",
            "_dom_classes": [],
            "description": "Computing statistics...: 100%",
            "_model_name": "FloatProgressModel",
            "bar_style": "success",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_5338e467fcfb4df28e4ce1127f4dba3c"
          }
        },
        "2dc78bf5fe0a4c02b75af9e5b2a6a0dc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_5e00e35ca6b44b27bf89fc898a6866de",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 1/1 [00:50&lt;00:00, 50.19s/ split]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_b0455799d03a49bd8a18b3b29fc5d93b"
          }
        },
        "d8070f6d74844ae197c0f8ef6a82c053": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "5338e467fcfb4df28e4ce1127f4dba3c": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "5e00e35ca6b44b27bf89fc898a6866de": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "b0455799d03a49bd8a18b3b29fc5d93b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "959cbdf31f01479f8c481d302c5aa028": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "state": {
            "_view_name": "HBoxView",
            "_dom_classes": [],
            "_model_name": "HBoxModel",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "box_style": "",
            "layout": "IPY_MODEL_c29be1c283ee4688a5e67c77250e98fb",
            "_model_module": "@jupyter-widgets/controls",
            "children": [
              "IPY_MODEL_71572360a7604deaa84dae8aa276e62d",
              "IPY_MODEL_9db9bbf37d07484eb7ff5042b784c7ab"
            ]
          }
        },
        "c29be1c283ee4688a5e67c77250e98fb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "71572360a7604deaa84dae8aa276e62d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "state": {
            "_view_name": "ProgressView",
            "style": "IPY_MODEL_f9d2d476fc4441c7906ca57c66179cc1",
            "_dom_classes": [],
            "description": "",
            "_model_name": "FloatProgressModel",
            "bar_style": "info",
            "max": 1,
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": 1,
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "orientation": "horizontal",
            "min": 0,
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_28f5b3d697894eb3a44b36064b860a13"
          }
        },
        "9db9bbf37d07484eb7ff5042b784c7ab": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "state": {
            "_view_name": "HTMLView",
            "style": "IPY_MODEL_0daae13174534fa4befd66f307a27aec",
            "_dom_classes": [],
            "description": "",
            "_model_name": "HTMLModel",
            "placeholder": "​",
            "_view_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "value": " 27558/? [00:22&lt;00:00, 1440.71 examples/s]",
            "_view_count": null,
            "_view_module_version": "1.5.0",
            "description_tooltip": null,
            "_model_module": "@jupyter-widgets/controls",
            "layout": "IPY_MODEL_a81baac7a57f4b3ab06c0906878d0e91"
          }
        },
        "f9d2d476fc4441c7906ca57c66179cc1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "ProgressStyleModel",
            "description_width": "initial",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "bar_color": null,
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "28f5b3d697894eb3a44b36064b860a13": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        },
        "0daae13174534fa4befd66f307a27aec": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_view_name": "StyleView",
            "_model_name": "DescriptionStyleModel",
            "description_width": "",
            "_view_module": "@jupyter-widgets/base",
            "_model_module_version": "1.5.0",
            "_view_count": null,
            "_view_module_version": "1.2.0",
            "_model_module": "@jupyter-widgets/controls"
          }
        },
        "a81baac7a57f4b3ab06c0906878d0e91": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "state": {
            "_view_name": "LayoutView",
            "grid_template_rows": null,
            "right": null,
            "justify_content": null,
            "_view_module": "@jupyter-widgets/base",
            "overflow": null,
            "_model_module_version": "1.2.0",
            "_view_count": null,
            "flex_flow": null,
            "width": null,
            "min_width": null,
            "border": null,
            "align_items": null,
            "bottom": null,
            "_model_module": "@jupyter-widgets/base",
            "top": null,
            "grid_column": null,
            "overflow_y": null,
            "overflow_x": null,
            "grid_auto_flow": null,
            "grid_area": null,
            "grid_template_columns": null,
            "flex": null,
            "_model_name": "LayoutModel",
            "justify_items": null,
            "grid_row": null,
            "max_height": null,
            "align_content": null,
            "visibility": null,
            "align_self": null,
            "height": null,
            "min_height": null,
            "padding": null,
            "grid_auto_rows": null,
            "grid_gap": null,
            "max_width": null,
            "order": null,
            "_view_module_version": "1.2.0",
            "grid_template_areas": null,
            "object_position": null,
            "object_fit": null,
            "grid_auto_columns": null,
            "margin": null,
            "display": null,
            "left": null
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/nestorsgarzonc/Data-Science-Proyects/blob/master/malaria_detection.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5wSiGqH_C8LK",
        "colab_type": "text"
      },
      "source": [
        "### Import Libraries"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NQ601KPyCyNi",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "import tensorflow as tf\n",
        "import tensorflow_datasets as tfds"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "mJJEMIokDmF2",
        "colab_type": "text"
      },
      "source": [
        "### Load data"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Rzlh-o93Dj8x",
        "colab_type": "code",
        "outputId": "07bf5b32-ac0d-4d47-e06a-562f5c00cf53",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 567
        }
      },
      "source": [
        "ds_builder=tfds.builder(name='malaria')\n",
        "print(ds_builder.info)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "tfds.core.DatasetInfo(\n",
            "    name='malaria',\n",
            "    version=1.0.0,\n",
            "    description='The Malaria dataset contains a total of 27,558 cell images\n",
            "with equal instances of parasitized and uninfected cells from the thin blood \n",
            "smear slide images of segmented cells.',\n",
            "    homepage='https://lhncbc.nlm.nih.gov/publication/pub9932',\n",
            "    features=FeaturesDict({\n",
            "        'image': Image(shape=(None, None, 3), dtype=tf.uint8),\n",
            "        'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=2),\n",
            "    }),\n",
            "    total_num_examples=27558,\n",
            "    splits={\n",
            "        'train': 27558,\n",
            "    },\n",
            "    supervised_keys=('image', 'label'),\n",
            "    citation=\"\"\"@article{rajaraman2018pre,\n",
            "      title={Pre-trained convolutional neural networks as feature extractors toward \n",
            "      improved malaria parasite detection in thin blood smear images},\n",
            "      author={Rajaraman, Sivaramakrishnan and Antani, Sameer K and Poostchi, Mahdieh\n",
            "      and Silamut, Kamolrat and Hossain, Md A and Maude, Richard J and Jaeger, \n",
            "      Stefan and Thoma, George R},\n",
            "      journal={PeerJ},\n",
            "      volume={6},\n",
            "      pages={e4568},\n",
            "      year={2018},\n",
            "      publisher={PeerJ Inc.}\n",
            "    }\"\"\",\n",
            "    redistribution_info=,\n",
            ")\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sdz_O20pDuxe",
        "colab_type": "code",
        "outputId": "0eaf73b2-56c4-4a20-f785-06d4133451c2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 410,
          "referenced_widgets": [
            "f79e1dee556544cfb875121aef27c376",
            "3e57e7591462419395966022a00f3492",
            "363adaa5bf3643d8a925453c60e20b77",
            "983fb7768bfc4af1ab2784cb97246931",
            "2ed1c4121841437f8317486bbd8d02f6",
            "46e7c84750f146419a7b7e94d67263c8",
            "3d6fbd4aeb4c4762a1ae335d366a5503",
            "1b718f447be848248c8c93c9c69f6244",
            "e1d2814c9b72448f987b71f558a2f4cb",
            "534d52c5aeb844a7b86f4dadbeb5b14f",
            "4e415db71c68415a9110e0721681e8d4",
            "89f9120489cd4ac8b8c84a65c7b38822",
            "5e46c33c819041cfa50376e81254ac39",
            "50ab3fb0afeb4261983383f8b95529e6",
            "d45b48af493840f4b767d09e97fb6c55",
            "bbfe2e6d491b4cf1864f1388a32b8aa1",
            "ea80d6e88f6b493aa0feb35fcaa70124",
            "4e7df5b9fafa4ac388da1be19dcae056",
            "bcf4803447d84894b8deedfe64c85c65",
            "353701ba92fa419b90a9c88d331cc807",
            "01583e8bf90d4fcabf4da636b586922e",
            "cb600d99f4a942a694716c56e3c3e5a0",
            "06e89eb94cc64fe2aafc413a925bb854",
            "bb17062811234f44b5e9ba875da45e77",
            "0cd1d3ba1dd74d2f98445758392db6ec",
            "e182dd7451b7464f99ea578854597525",
            "645e22283c8c4d9081b901dba6185169",
            "78e3b6f820e54bd29df7a1e3102b54bd",
            "8d6f1f2d1e7743f7997f6c4a67bc5cb7",
            "0b76d70cc7b84e2ebeeea210ae189ac5",
            "8558ea86176f4f77922a9b767b477a30",
            "1c7963b3fddb4c6fac1de936488f395a",
            "73a0ee47ab2a4f9f9fe160d901a3aa47",
            "3393b45012be4a99aa69f00507d654c1",
            "b3c1dc28d48342eeabcc10c847023ebd",
            "cedc522d27f24be58fea368900bfe447",
            "d37c1169e9634f98ad0bca7e3293633d",
            "ec5c8f36f679419db79f28a32ea947de",
            "e956e91be6e94fcaa6728b6e62158414",
            "8f102d253fe841a09af03abf90a63967",
            "8cc0c81d1f54464eb4668fe3e64aeeba",
            "69e0bbf660c14691873bd2809f290be7",
            "3e98141c909d4463835c530846502760",
            "2dc78bf5fe0a4c02b75af9e5b2a6a0dc",
            "d8070f6d74844ae197c0f8ef6a82c053",
            "5338e467fcfb4df28e4ce1127f4dba3c",
            "5e00e35ca6b44b27bf89fc898a6866de",
            "b0455799d03a49bd8a18b3b29fc5d93b",
            "959cbdf31f01479f8c481d302c5aa028",
            "c29be1c283ee4688a5e67c77250e98fb",
            "71572360a7604deaa84dae8aa276e62d",
            "9db9bbf37d07484eb7ff5042b784c7ab",
            "f9d2d476fc4441c7906ca57c66179cc1",
            "28f5b3d697894eb3a44b36064b860a13",
            "0daae13174534fa4befd66f307a27aec",
            "a81baac7a57f4b3ab06c0906878d0e91"
          ]
        }
      },
      "source": [
        "ds=tfds.load(name='malaria', split='train')\n",
        "ds_numpy=tfds.as_numpy(ds)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "\u001b[1mDownloading and preparing dataset malaria/1.0.0 (download: 337.08 MiB, generated: Unknown size, total: 337.08 MiB) to /root/tensorflow_datasets/malaria/1.0.0...\u001b[0m\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "f79e1dee556544cfb875121aef27c376",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Dl Completed...', max=1.0, style=Progre…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "e1d2814c9b72448f987b71f558a2f4cb",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Dl Size...', max=1.0, style=ProgressSty…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "ea80d6e88f6b493aa0feb35fcaa70124",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', description='Extraction completed...', max=1.0, styl…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/urllib3/connectionpool.py:847: InsecureRequestWarning: Unverified HTTPS request is being made. Adding certificate verification is strongly advised. See: https://urllib3.readthedocs.io/en/latest/advanced-usage.html#ssl-warnings\n",
            "  InsecureRequestWarning)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "stream",
          "text": [
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "0cd1d3ba1dd74d2f98445758392db6ec",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\rShuffling and writing examples to /root/tensorflow_datasets/malaria/1.0.0.incompleteWWR3ZO/malaria-train.tfrecord\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "73a0ee47ab2a4f9f9fe160d901a3aa47",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, max=27558.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "8cc0c81d1f54464eb4668fe3e64aeeba",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=0.0, description='Computing statistics...', max=1.0, style=ProgressStyle(de…"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\r"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "959cbdf31f01479f8c481d302c5aa028",
              "version_minor": 0,
              "version_major": 2
            },
            "text/plain": [
              "HBox(children=(FloatProgress(value=1.0, bar_style='info', max=1.0), HTML(value='')))"
            ]
          },
          "metadata": {
            "tags": []
          }
        },
        {
          "output_type": "stream",
          "text": [
            "\r\n",
            "\u001b[1mDataset malaria downloaded and prepared to /root/tensorflow_datasets/malaria/1.0.0. Subsequent calls will reuse this data.\u001b[0m\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "EsMXjSFMjUv7",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x = []\n",
        "y = []\n",
        "for ex in ds_numpy:\n",
        "  # `{'image': np.array(shape=(None, None, 1)), 'label': np.array(shape=())}`\n",
        "  x.append(ex['image'])\n",
        "  y.append(ex['label'])\n",
        "\n",
        "# And transform them into numpy arrays\n",
        "x, y = np.asarray(x), np.asarray(y)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jQYEh91hkBej",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Liberar memoria\n",
        "del ds_numpy\n",
        "del ds"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YbO5jkuWkqy-",
        "colab_type": "code",
        "outputId": "5edccafd-fe42-4d8b-faf5-2d9672087d4e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "print(x.shape, y.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(27558,) (27558,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bCp5xKPRmMhv",
        "colab_type": "text"
      },
      "source": [
        "### EDA (Exploratory Data Analysis)"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zc2RHDknktz4",
        "colab_type": "code",
        "outputId": "919d0ee3-7b57-4b38-f5eb-0942c1550f13",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#Imagenes\n",
        "x.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(27558,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7rwx8ILgmXYm",
        "colab_type": "code",
        "outputId": "b3ffc729-56c0-4586-8791-d0ed0a47d233",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#Labels 0 or 1\n",
        "y.shape"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(27558,)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "wo5Ej5FQmYEy",
        "colab_type": "code",
        "outputId": "bcba1eb5-4f19-42c2-b4d2-11c8744a653e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#Mirar cuantos datos unicos hay\n",
        "print(np.unique(y, return_counts=True))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(array([0, 1]), array([13779, 13779]))\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ec3Zz5bvm2M8",
        "colab_type": "text"
      },
      "source": [
        "El dataset esta balanceado\n",
        "\n",
        "Ahora imprimir las imagenes"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rAJjpQlFmgBF",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Mostrar aleatoriamente 6 imagenes de las clases deseadas\n",
        "def plot_imgs(x, y, class_num):\n",
        "  plt.figure(figsize=(12,12))\n",
        "  for i in range(6):\n",
        "    plt.subplot(1,6,i+1)\n",
        "    random_ind=np.random.randint(0, len(x))\n",
        "\n",
        "    #Intentar hasta volver a asignar un ejemplo de la clase deseada\n",
        "    while y[random_ind]!=class_num:\n",
        "      random_ind=np.random.randint(0, len(x))\n",
        "\n",
        "    plt.imshow(x[random_ind])\n",
        "    plt.title(str(class_num))\n",
        "    plt.tight_layout()\n",
        "  plt.show()"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fLnv8Ureov3e",
        "colab_type": "text"
      },
      "source": [
        "Imagenes con patologia"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GHyym9nJnumn",
        "colab_type": "code",
        "outputId": "527c9459-2595-4d66-b609-c5471e3defb1",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 154
        }
      },
      "source": [
        "plot_imgs(x,y,0)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x864 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vUUJ0gq8o0Cs",
        "colab_type": "text"
      },
      "source": [
        "Imagenes sin patologia"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "rDd7kmtGny7I",
        "colab_type": "code",
        "outputId": "0295ce5c-6e74-41e6-b9d8-e766e3572ef2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 153
        }
      },
      "source": [
        "plot_imgs(x,y,1)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x864 with 6 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hCC6o0OKpWZ_",
        "colab_type": "text"
      },
      "source": [
        "### Data preprocessing\n",
        " \n",
        "\n",
        "1.   train test split\n",
        "2.   standarize\n",
        "3.   resize\n",
        "4.   one-shot enconding\n",
        "5.   pixel value normalization\n",
        "\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qrfFVQwVo1gY",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3, random_state=666)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1qfVhcMEqJmW",
        "colab_type": "code",
        "outputId": "d2448642-b823-4025-fcbd-bb0e6c82b012",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 123
        }
      },
      "source": [
        "print('train')\n",
        "print(x_train.shape, y_train.shape)\n",
        "print(np.unique(y_train, return_counts=True))\n",
        "\n",
        "print('test')\n",
        "print(x_test.shape, y_test.shape)\n",
        "print(np.unique(y_test, return_counts=True))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train\n",
            "(19290,) (19290,)\n",
            "(array([0, 1]), array([9614, 9676]))\n",
            "test\n",
            "(8268,) (8268,)\n",
            "(array([0, 1]), array([4165, 4103]))\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "eWncV2mVrebb",
        "colab_type": "text"
      },
      "source": [
        "Como se puede ver los sets de datos no estan balanceados, para ello usar <strong>STRATIFICATION</strong>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H2IMpe6LqWpl",
        "colab_type": "code",
        "outputId": "3afc29b5-dfe1-40e8-ce02-89bbf08dc2c5",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 123
        }
      },
      "source": [
        "x_train, x_test, y_train, y_test=train_test_split(x,y, stratify=y, test_size=0.3, random_state=666)\n",
        "\n",
        "print('train')\n",
        "print(x_train.shape, y_train.shape)\n",
        "print(np.unique(y_train, return_counts=True))\n",
        "\n",
        "print('test')\n",
        "print(x_test.shape, y_test.shape)\n",
        "print(np.unique(y_test, return_counts=True))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "train\n",
            "(19290,) (19290,)\n",
            "(array([0, 1]), array([9645, 9645]))\n",
            "test\n",
            "(8268,) (8268,)\n",
            "(array([0, 1]), array([4134, 4134]))\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-7_iTEcgsGD2",
        "colab_type": "text"
      },
      "source": [
        "Check mean, min, max"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mMDIR5BHr3-D",
        "colab_type": "code",
        "outputId": "2bee2bb1-d208-45f5-cb49-6c332a30b467",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        }
      },
      "source": [
        "#h=height\n",
        "#w=width\n",
        "h_size=[]\n",
        "w_size=[]\n",
        "\n",
        "for img in x_test:\n",
        "  h_size.append(img.shape[0])\n",
        "  w_size.append(img.shape[1])\n",
        "\n",
        "h_size, w_size=np.asarray(h_size), np.asarray(w_size)\n",
        "print('Height')\n",
        "print(np.min(h_size), np.mean(h_size), np.max(h_size))\n",
        "print('Width')\n",
        "print(np.min(w_size), np.mean(w_size), np.max(w_size))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Height\n",
            "49 132.8178519593614 340\n",
            "Width\n",
            "46 132.47206095791 364\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mcggh9BPswZl",
        "colab_type": "code",
        "outputId": "4cfcdb7a-e8c2-4f16-a2bb-8669ab22a6bf",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34
        }
      },
      "source": [
        "#resize img ya que estan en diferentes tamaños\n",
        "import cv2 #paquete computer vision, tratamiento imagenes\n",
        "size=50\n",
        "\n",
        "x_train_input=[]\n",
        "for i in range(len(x_train)):\n",
        "  img=cv2.resize(x_train[i], (size, size))\n",
        "  x_train_input.append(img)\n",
        "\n",
        "x_train_input=np.asarray(x_train_input)\n",
        "y_train_input=np.copy(y_train)\n",
        "\n",
        "print(x_train_input.shape, y_train_input.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(19290, 50, 50, 3) (19290,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Ju-y5Tv2vXp-",
        "colab_type": "text"
      },
      "source": [
        "One hot encoding: poner numeero de clases en un vector por cada clase\n",
        "\n",
        "Ejemplo, infectado o no infectado se puede representar como [infectado, no infectado]\n",
        "Para n categorias seria: [0, 0, n_categorias]"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "hdP0yD_VtpZx",
        "colab_type": "code",
        "outputId": "f6f5e270-861e-489a-c057-b2d9d7fb3035",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        }
      },
      "source": [
        "#data to categorical, modelo general\n",
        "import keras\n",
        "num_classes=2\n",
        "y_train_input=keras.utils.to_categorical(y_train_input, num_classes)\n",
        "\n",
        "print(y_train_input.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "(19290, 2)\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "stream",
          "text": [
            "Using TensorFlow backend.\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Z-Lhrkm7wErN",
        "colab_type": "text"
      },
      "source": [
        "Normalize the data\n",
        "En este caso como son imagenes y los valores van desde 0 hasta 255, hay que dividir el tensor entre 255 aprovechando el broadcasting de python"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GpxbgGYauoRr",
        "colab_type": "code",
        "outputId": "9837d563-be4a-4f53-8f32-acf4e69bc456",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52
        }
      },
      "source": [
        "print(np.min(x_train_input), np.max(x_train_input))\n",
        "\n",
        "x_train_input = x_train_input*1.0/255.\n",
        "\n",
        "print(np.min(x_train_input), np.max(x_train_input))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "0 255\n",
            "0.0 1.0\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "IEanUOyzwf_3",
        "colab_type": "code",
        "outputId": "cdc5f768-3682-4c91-fbbc-df059f3d55c8",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        }
      },
      "source": [
        "#Data is ready\n",
        "print('Train')\n",
        "print(x_train_input.shape, y_train_input.shape)\n",
        "\n",
        "print('Test')\n",
        "print(x_test.shape, y_test.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train\n",
            "(19290, 50, 50, 3) (19290, 2)\n",
            "Test\n",
            "(8268,) (8268,)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6a50KmPjxlIX",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def preprocess_data(x_input, y_input, num_classes, size):\n",
        "  \"\"\"\n",
        "  This function do the same that we do before:\n",
        "  resize the data\n",
        "  normalize\n",
        "  data to categorical\n",
        "  \"\"\"\n",
        "  \n",
        "  x_output=[]\n",
        "  for i in range(len(x_input)):\n",
        "    img=cv2.resize(x_input[i], (size, size))\n",
        "    x_output.append(img)\n",
        "\n",
        "  x_output=np.asarray(x_output)\n",
        "  y_output=np.copy(y_input)\n",
        "\n",
        "  x_output=x_output*1.0/255.\n",
        "\n",
        "  y_output=keras.utils.to_categorical(y_output, num_classes=num_classes)\n",
        "  \n",
        "  return x_output, y_output"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gK3eLtmMyMJP",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "x_test_input, y_test_input = preprocess_data(x_test, y_test, num_classes, size)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mbq5Bbl2ytw0",
        "colab_type": "code",
        "outputId": "65bc6d17-5d2c-4c7d-d1ae-29dafe57fefd",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 87
        }
      },
      "source": [
        "#Data is ready\n",
        "print('Train')\n",
        "print(x_train_input.shape, y_train_input.shape)\n",
        "\n",
        "print('Test')\n",
        "print(x_test_input.shape, y_test_input.shape)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train\n",
            "(19290, 50, 50, 3) (19290, 2)\n",
            "Test\n",
            "(8268, 50, 50, 3) (8268, 2)\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E_IYcFpCy74s",
        "colab_type": "text"
      },
      "source": [
        "### Deep learning\n",
        "Apply cnn model"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "heb-rZUuzGBL",
        "colab_type": "text"
      },
      "source": [
        "## Base model\n",
        "\n",
        "*    build\n",
        "*    train\n",
        "*    evaluate"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uZ2N3icGy2Yl",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.layers import Dense, Conv2D\n",
        "from keras.layers import Flatten\n",
        "from keras.layers import MaxPooling2D, GlobalAveragePooling2D\n",
        "from keras.layers import Activation\n",
        "from keras.layers import BatchNormalization\n",
        "from keras.layers import Dropout\n",
        "from keras.models import Sequential\n",
        "from keras import backend as K\n",
        "from keras import optimizers"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "QfAg7e68ztCQ",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def CNNbuild(height, width, classes, channels):\n",
        "  model=Sequential()\n",
        "  inputShape=(height, width, channels)\n",
        "\n",
        "  #Feature extractor\n",
        "  #Conv2D(#filters, (size_filter, size_filter))\n",
        "  model.add(Conv2D(32, (3,3), activation='relu', input_shape=inputShape))\n",
        "  model.add(MaxPooling2D(2,2))\n",
        "  #dropout: desactiva ciertas neuronas con x probabilidad\n",
        "  model.add(Dropout(0.2))\n",
        "\n",
        "  model.add(Conv2D(32, (3,3), activation='relu'))\n",
        "  model.add(MaxPooling2D(2,2))\n",
        "  model.add(Dropout(0.2))\n",
        "\n",
        "  model.add(Conv2D(32, (3,3), activation='relu'))\n",
        "  model.add(MaxPooling2D(2,2))\n",
        "  model.add(Dropout(0.2))\n",
        "\n",
        "  #Fully connected FC\n",
        "  model.add(Flatten())\n",
        "\n",
        "  model.add(Dense(512, activation='relu'))\n",
        "  model.add(Dropout(0.5))\n",
        "  model.add(Dense(classes, activation='softmax'))\n",
        "\n",
        "  return model"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SAZW--_i3b1H",
        "colab_type": "code",
        "outputId": "8b34f144-41fe-4fce-df7f-ee04f6e005c2",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 638
        }
      },
      "source": [
        "height=x_train_input.shape[1]\n",
        "width=x_train_input.shape[2]\n",
        "channels=x_train_input.shape[3]\n",
        "print(height, width, channels)\n",
        "\n",
        "model=CNNbuild(height, width, num_classes, channels)\n",
        "\n",
        "model.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "50 50 3\n",
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_1 (Conv2D)            (None, 48, 48, 32)        896       \n",
            "_________________________________________________________________\n",
            "max_pooling2d_1 (MaxPooling2 (None, 24, 24, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout_1 (Dropout)          (None, 24, 24, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_2 (Conv2D)            (None, 22, 22, 32)        9248      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_2 (MaxPooling2 (None, 11, 11, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout_2 (Dropout)          (None, 11, 11, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_3 (Conv2D)            (None, 9, 9, 32)          9248      \n",
            "_________________________________________________________________\n",
            "max_pooling2d_3 (MaxPooling2 (None, 4, 4, 32)          0         \n",
            "_________________________________________________________________\n",
            "dropout_3 (Dropout)          (None, 4, 4, 32)          0         \n",
            "_________________________________________________________________\n",
            "flatten_1 (Flatten)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_1 (Dense)              (None, 512)               262656    \n",
            "_________________________________________________________________\n",
            "dropout_4 (Dropout)          (None, 512)               0         \n",
            "_________________________________________________________________\n",
            "dense_2 (Dense)              (None, 2)                 1026      \n",
            "=================================================================\n",
            "Total params: 283,074\n",
            "Trainable params: 283,074\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "m8CjB6vU7gts",
        "colab_type": "text"
      },
      "source": [
        "## Model training"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "mVaCSsWV6BCG",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Hyperparameters\n",
        "epochs=20\n",
        "batch_size=32"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "MdFn_9cF7nkI",
        "colab_type": "code",
        "outputId": "d53b6bd9-4a56-4c81-846c-b7aee1702a0b",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 765
        }
      },
      "source": [
        "#Compile model\n",
        "model.compile(loss='categorical_crossentropy', optimizer='Adam', metrics=['accuracy'])\n",
        "h=model.fit(x_train_input, y_train_input, epochs=epochs, batch_size=batch_size, validation_data=(x_test_input, y_test_input))"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 19290 samples, validate on 8268 samples\n",
            "Epoch 1/20\n",
            "19290/19290 [==============================] - 12s 622us/step - loss: 0.4457 - accuracy: 0.7723 - val_loss: 0.1904 - val_accuracy: 0.9360\n",
            "Epoch 2/20\n",
            "19290/19290 [==============================] - 5s 281us/step - loss: 0.1805 - accuracy: 0.9430 - val_loss: 0.1660 - val_accuracy: 0.9473\n",
            "Epoch 3/20\n",
            "19290/19290 [==============================] - 5s 279us/step - loss: 0.1633 - accuracy: 0.9480 - val_loss: 0.1444 - val_accuracy: 0.9539\n",
            "Epoch 4/20\n",
            "19290/19290 [==============================] - 5s 280us/step - loss: 0.1550 - accuracy: 0.9503 - val_loss: 0.1508 - val_accuracy: 0.9538\n",
            "Epoch 5/20\n",
            "19290/19290 [==============================] - 5s 280us/step - loss: 0.1479 - accuracy: 0.9541 - val_loss: 0.1412 - val_accuracy: 0.9548\n",
            "Epoch 6/20\n",
            "19290/19290 [==============================] - 5s 278us/step - loss: 0.1465 - accuracy: 0.9543 - val_loss: 0.1456 - val_accuracy: 0.9551\n",
            "Epoch 7/20\n",
            "19290/19290 [==============================] - 5s 279us/step - loss: 0.1456 - accuracy: 0.9529 - val_loss: 0.1417 - val_accuracy: 0.9557\n",
            "Epoch 8/20\n",
            "19290/19290 [==============================] - 5s 277us/step - loss: 0.1366 - accuracy: 0.9551 - val_loss: 0.1436 - val_accuracy: 0.9562\n",
            "Epoch 9/20\n",
            "19290/19290 [==============================] - 5s 281us/step - loss: 0.1383 - accuracy: 0.9544 - val_loss: 0.1522 - val_accuracy: 0.9527\n",
            "Epoch 10/20\n",
            "19290/19290 [==============================] - 5s 278us/step - loss: 0.1351 - accuracy: 0.9549 - val_loss: 0.1540 - val_accuracy: 0.9523\n",
            "Epoch 11/20\n",
            "19290/19290 [==============================] - 5s 281us/step - loss: 0.1312 - accuracy: 0.9568 - val_loss: 0.1405 - val_accuracy: 0.9544\n",
            "Epoch 12/20\n",
            "19290/19290 [==============================] - 5s 278us/step - loss: 0.1319 - accuracy: 0.9563 - val_loss: 0.1390 - val_accuracy: 0.9548\n",
            "Epoch 13/20\n",
            "19290/19290 [==============================] - 5s 278us/step - loss: 0.1302 - accuracy: 0.9564 - val_loss: 0.1442 - val_accuracy: 0.9532\n",
            "Epoch 14/20\n",
            "19290/19290 [==============================] - 5s 283us/step - loss: 0.1289 - accuracy: 0.9574 - val_loss: 0.1610 - val_accuracy: 0.9502\n",
            "Epoch 15/20\n",
            "19290/19290 [==============================] - 6s 294us/step - loss: 0.1269 - accuracy: 0.9583 - val_loss: 0.1402 - val_accuracy: 0.9530\n",
            "Epoch 16/20\n",
            "19290/19290 [==============================] - 6s 291us/step - loss: 0.1209 - accuracy: 0.9602 - val_loss: 0.1358 - val_accuracy: 0.9532\n",
            "Epoch 17/20\n",
            "19290/19290 [==============================] - 5s 283us/step - loss: 0.1209 - accuracy: 0.9586 - val_loss: 0.1397 - val_accuracy: 0.9548\n",
            "Epoch 18/20\n",
            "19290/19290 [==============================] - 5s 281us/step - loss: 0.1207 - accuracy: 0.9594 - val_loss: 0.1377 - val_accuracy: 0.9540\n",
            "Epoch 19/20\n",
            "19290/19290 [==============================] - 5s 279us/step - loss: 0.1185 - accuracy: 0.9598 - val_loss: 0.1305 - val_accuracy: 0.9572\n",
            "Epoch 20/20\n",
            "19290/19290 [==============================] - 5s 279us/step - loss: 0.1161 - accuracy: 0.9604 - val_loss: 0.1322 - val_accuracy: 0.9560\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bSK5o4jBRAtA",
        "colab_type": "text"
      },
      "source": [
        "### Model evaluation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2y6y1vEe8ZRZ",
        "colab_type": "code",
        "outputId": "1155fd09-2339-4ae2-f7c1-2e4a1433bb11",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 105
        }
      },
      "source": [
        "from sklearn.metrics import balanced_accuracy_score, accuracy_score, f1_score, confusion_matrix\n",
        "\n",
        "y_pred=model.predict(x_test_input)\n",
        "#Use argmax\n",
        "y_pred=np.argmax(y_pred, axis=-1)\n",
        "\n",
        "acc= accuracy_score(y_test, y_pred)\n",
        "bac=balanced_accuracy_score(y_test, y_pred)\n",
        "f1=f1_score(y_test, y_pred)\n",
        "cm=confusion_matrix(y_test, y_pred)\n",
        "\n",
        "print('Accuracy: {}'.format(str(acc*100)+'%'))\n",
        "print('BAC: {}'.format(str(bac*100)+'%'))\n",
        "print('F1 score: {}'.format(str(f1*100)+'%'))\n",
        "\n",
        "print(cm)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Accuracy: 95.59748427672956%\n",
            "BAC: 95.59748427672956%\n",
            "F1 score: 95.67284831193533%\n",
            "[[3880  254]\n",
            " [ 110 4024]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "SkZhgBfwRCDm",
        "colab_type": "code",
        "outputId": "b5befed7-4afa-404f-a37e-ea9ae7f644c9",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 457
        }
      },
      "source": [
        "#Plot confusion matrix\n",
        "import pandas as pd\n",
        "import seaborn as sn\n",
        "\n",
        "cm_df=pd.DataFrame(cm, dtype=int)\n",
        "plt.figure(figsize=(10,7))\n",
        "sn.set(font_scale=1.4)\n",
        "sn.heatmap(cm_df, annot=True, annot_kws={'size':16}, fmt='g', cmap='Blues')"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f792e0258d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 33
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fXR7HW_KUW_O",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "#Find missclassified samples\n",
        "ind_error=[]\n",
        "for i in range(len(y_test)):\n",
        "  if y_test[i] != y_pred[i]:\n",
        "    ind_error.append(i)"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "JjNarHgfVHwf",
        "colab_type": "code",
        "outputId": "1df6b6f8-8d69-4b4a-da9a-659e640c0595",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 314
        }
      },
      "source": [
        "plt.imshow(x_test[ind_error[0]])\n",
        "plt.title('real class {} \\nPredicted Class: {}'.format(y_test[ind_error[0]], y_pred[ind_error[0]]))\n",
        "plt.show()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rmPB9JUXWiGk",
        "colab_type": "text"
      },
      "source": [
        "### Base model 2"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sIt7K6l7VfSw",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "def CNNbuild2(height, width, classes, channels):\n",
        "  model=keras.Sequential()\n",
        "\n",
        "  model.add(Conv2D(32, (3,3), padding='same', input_shape=(height, width, 3)))\n",
        "  model.add(Activation('relu'))\n",
        "  model.add(Conv2D(32, (3,3)))\n",
        "  model.add(Activation('relu'))\n",
        "  model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "  model.add(Dropout(0.25))\n",
        "\n",
        "  \n",
        "  model.add(Conv2D(64, (3,3), padding='same'))\n",
        "  model.add(Activation('relu'))\n",
        "  model.add(Conv2D(64, (3,3)))\n",
        "  model.add(Activation('relu'))\n",
        "  model.add(MaxPooling2D(pool_size=(2,2)))\n",
        "  model.add(Dropout(0.25))\n",
        "\n",
        "  model.add(Flatten())\n",
        "  model.add(Dropout(0.5))\n",
        "  model.add(Dense(64))\n",
        "  model.add(Activation('relu'))\n",
        "  model.add(Dropout(0.5))\n",
        "  model.add(Dense(num_classes))\n",
        "  model.add(Activation('softmax'))\n",
        "  \n",
        "  return model\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fHe3cw7NalDe",
        "colab_type": "text"
      },
      "source": [
        "# Create the network"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Cj1bWf7PXmJe",
        "colab_type": "code",
        "outputId": "09c6bbe6-d64b-4003-f079-1abee49e11ba",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 852
        }
      },
      "source": [
        "height=x_train_input.shape[1]\n",
        "width=x_train_input.shape[2]\n",
        "channels=x_train_input.shape[3]\n",
        "print(height,width,channels)\n",
        "\n",
        "model_2=CNNbuild2(height, width, num_classes, channels)\n",
        "model_2.summary()"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "50 50 3\n",
            "Model: \"sequential_3\"\n",
            "_________________________________________________________________\n",
            "Layer (type)                 Output Shape              Param #   \n",
            "=================================================================\n",
            "conv2d_5 (Conv2D)            (None, 50, 50, 32)        896       \n",
            "_________________________________________________________________\n",
            "activation_1 (Activation)    (None, 50, 50, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_6 (Conv2D)            (None, 48, 48, 32)        9248      \n",
            "_________________________________________________________________\n",
            "activation_2 (Activation)    (None, 48, 48, 32)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_4 (MaxPooling2 (None, 24, 24, 32)        0         \n",
            "_________________________________________________________________\n",
            "dropout_5 (Dropout)          (None, 24, 24, 32)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_7 (Conv2D)            (None, 24, 24, 64)        18496     \n",
            "_________________________________________________________________\n",
            "activation_3 (Activation)    (None, 24, 24, 64)        0         \n",
            "_________________________________________________________________\n",
            "conv2d_8 (Conv2D)            (None, 22, 22, 64)        36928     \n",
            "_________________________________________________________________\n",
            "activation_4 (Activation)    (None, 22, 22, 64)        0         \n",
            "_________________________________________________________________\n",
            "max_pooling2d_5 (MaxPooling2 (None, 11, 11, 64)        0         \n",
            "_________________________________________________________________\n",
            "dropout_6 (Dropout)          (None, 11, 11, 64)        0         \n",
            "_________________________________________________________________\n",
            "flatten_2 (Flatten)          (None, 7744)              0         \n",
            "_________________________________________________________________\n",
            "dropout_7 (Dropout)          (None, 7744)              0         \n",
            "_________________________________________________________________\n",
            "dense_3 (Dense)              (None, 64)                495680    \n",
            "_________________________________________________________________\n",
            "activation_5 (Activation)    (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dropout_8 (Dropout)          (None, 64)                0         \n",
            "_________________________________________________________________\n",
            "dense_4 (Dense)              (None, 2)                 130       \n",
            "_________________________________________________________________\n",
            "activation_6 (Activation)    (None, 2)                 0         \n",
            "=================================================================\n",
            "Total params: 561,378\n",
            "Trainable params: 561,378\n",
            "Non-trainable params: 0\n",
            "_________________________________________________________________\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-Y4KFlwNbkoM",
        "colab_type": "text"
      },
      "source": [
        "# compile the model and train"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "sEDUoybubYcP",
        "colab_type": "code",
        "outputId": "b5bba181-f5f1-4d19-fbd0-e33dda968f03",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 765
        }
      },
      "source": [
        "model_2.compile(loss='categorical_crossentropy', optimizer='Adam', metrics=['accuracy'])\n",
        "h=model_2.fit(x_train_input, y_train_input, epochs=epochs, \n",
        "              batch_size=batch_size, \n",
        "              validation_data=(x_test_input, y_test_input)\n",
        "              )"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Train on 19290 samples, validate on 8268 samples\n",
            "Epoch 1/20\n",
            "19290/19290 [==============================] - 9s 470us/step - loss: 0.6534 - accuracy: 0.5975 - val_loss: 0.3169 - val_accuracy: 0.8985\n",
            "Epoch 2/20\n",
            "19290/19290 [==============================] - 9s 445us/step - loss: 0.5368 - accuracy: 0.6901 - val_loss: 0.1808 - val_accuracy: 0.9253\n",
            "Epoch 3/20\n",
            "19290/19290 [==============================] - 9s 443us/step - loss: 0.1884 - accuracy: 0.9330 - val_loss: 0.1716 - val_accuracy: 0.9083\n",
            "Epoch 4/20\n",
            "19290/19290 [==============================] - 9s 445us/step - loss: 0.1699 - accuracy: 0.9439 - val_loss: 0.1586 - val_accuracy: 0.9519\n",
            "Epoch 5/20\n",
            "19290/19290 [==============================] - 9s 449us/step - loss: 0.1596 - accuracy: 0.9510 - val_loss: 0.1380 - val_accuracy: 0.9523\n",
            "Epoch 6/20\n",
            "19290/19290 [==============================] - 9s 444us/step - loss: 0.1494 - accuracy: 0.9528 - val_loss: 0.1607 - val_accuracy: 0.9450\n",
            "Epoch 7/20\n",
            "19290/19290 [==============================] - 9s 441us/step - loss: 0.1457 - accuracy: 0.9540 - val_loss: 0.1435 - val_accuracy: 0.9531\n",
            "Epoch 8/20\n",
            "19290/19290 [==============================] - 9s 447us/step - loss: 0.1428 - accuracy: 0.9533 - val_loss: 0.1357 - val_accuracy: 0.9550\n",
            "Epoch 9/20\n",
            "19290/19290 [==============================] - 9s 442us/step - loss: 0.1406 - accuracy: 0.9549 - val_loss: 0.1476 - val_accuracy: 0.9526\n",
            "Epoch 10/20\n",
            "19290/19290 [==============================] - 9s 441us/step - loss: 0.1412 - accuracy: 0.9564 - val_loss: 0.1359 - val_accuracy: 0.9548\n",
            "Epoch 11/20\n",
            "19290/19290 [==============================] - 9s 444us/step - loss: 0.1387 - accuracy: 0.9556 - val_loss: 0.1347 - val_accuracy: 0.9542\n",
            "Epoch 12/20\n",
            "19290/19290 [==============================] - 8s 440us/step - loss: 0.1327 - accuracy: 0.9595 - val_loss: 0.1511 - val_accuracy: 0.9500\n",
            "Epoch 13/20\n",
            "19290/19290 [==============================] - 8s 440us/step - loss: 0.1290 - accuracy: 0.9580 - val_loss: 0.1268 - val_accuracy: 0.9550\n",
            "Epoch 14/20\n",
            "19290/19290 [==============================] - 9s 446us/step - loss: 0.1313 - accuracy: 0.9578 - val_loss: 0.1471 - val_accuracy: 0.9527\n",
            "Epoch 15/20\n",
            "19290/19290 [==============================] - 9s 443us/step - loss: 0.1263 - accuracy: 0.9589 - val_loss: 0.1459 - val_accuracy: 0.9482\n",
            "Epoch 16/20\n",
            "19290/19290 [==============================] - 9s 442us/step - loss: 0.1252 - accuracy: 0.9587 - val_loss: 0.1278 - val_accuracy: 0.9555\n",
            "Epoch 17/20\n",
            "19290/19290 [==============================] - 8s 438us/step - loss: 0.1245 - accuracy: 0.9600 - val_loss: 0.1468 - val_accuracy: 0.9530\n",
            "Epoch 18/20\n",
            "19290/19290 [==============================] - 9s 443us/step - loss: 0.1211 - accuracy: 0.9604 - val_loss: 0.1392 - val_accuracy: 0.9545\n",
            "Epoch 19/20\n",
            "19290/19290 [==============================] - 8s 439us/step - loss: 0.1207 - accuracy: 0.9593 - val_loss: 0.1246 - val_accuracy: 0.9543\n",
            "Epoch 20/20\n",
            "19290/19290 [==============================] - 8s 441us/step - loss: 0.1208 - accuracy: 0.9610 - val_loss: 0.1246 - val_accuracy: 0.9567\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "VL8gBPe6cuAd",
        "colab_type": "text"
      },
      "source": [
        "### Base model with augmentation"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "IlwlhTTwcxai",
        "colab_type": "text"
      },
      "source": [
        "# Build model"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "desio_KXcSLI",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "from keras.preprocessing.image import ImageDataGenerator"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KG5j8eYGdD6R",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        "train_datagen=ImageDataGenerator(\n",
        "    horizontal_flip=True,\n",
        "    width_shift_range=0.2,\n",
        "    height_shift_range=0.2,\n",
        "    fill_mode='nearest',\n",
        "    zoom_range=0.3,\n",
        "    rotation_range=30\n",
        ")\n",
        "test_datagen=ImageDataGenerator()\n",
        "\n",
        "train_generator=train_datagen.flow(x_train_input, y_train_input, batch_size=64,\n",
        "                                   shuffle=False)\n",
        "test_generator=train_datagen.flow(x_test_input, y_test_input, batch_size=64,\n",
        "                                   shuffle=False)\n"
      ],
      "execution_count": 0,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "2octPJQ9gNgC",
        "colab_type": "code",
        "outputId": "152602d0-b051-4685-8d1c-5c7c7d31ed6e",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 392
        }
      },
      "source": [
        "#Train model\n",
        "optim=optimizers.Adam(lr=0.001, decay=0.001/64)\n",
        "model.compile(loss='categorical_crossentropy', optimizer=optim, metrics=['accuracy'])\n",
        "h_aug=model.fit_generator(train_generator, steps_per_epoch=len(x_train)//64, epochs=10, validation_data=test_generator)"
      ],
      "execution_count": 0,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Epoch 1/10\n",
            "301/301 [==============================] - 20s 67ms/step - loss: 0.2702 - accuracy: 0.9034 - val_loss: 0.1198 - val_accuracy: 0.9109\n",
            "Epoch 2/10\n",
            "301/301 [==============================] - 20s 67ms/step - loss: 0.2583 - accuracy: 0.9060 - val_loss: 0.0642 - val_accuracy: 0.9153\n",
            "Epoch 3/10\n",
            "301/301 [==============================] - 20s 67ms/step - loss: 0.2570 - accuracy: 0.9075 - val_loss: 0.0323 - val_accuracy: 0.9118\n",
            "Epoch 4/10\n",
            "301/301 [==============================] - 20s 66ms/step - loss: 0.2544 - accuracy: 0.9081 - val_loss: 0.1875 - val_accuracy: 0.9065\n",
            "Epoch 5/10\n",
            "301/301 [==============================] - 20s 67ms/step - loss: 0.2544 - accuracy: 0.9086 - val_loss: 0.2091 - val_accuracy: 0.9111\n",
            "Epoch 6/10\n",
            "301/301 [==============================] - 20s 66ms/step - loss: 0.2532 - accuracy: 0.9080 - val_loss: 0.2051 - val_accuracy: 0.9110\n",
            "Epoch 7/10\n",
            "301/301 [==============================] - 20s 66ms/step - loss: 0.2495 - accuracy: 0.9116 - val_loss: 0.1659 - val_accuracy: 0.9159\n",
            "Epoch 8/10\n",
            "301/301 [==============================] - 20s 66ms/step - loss: 0.2456 - accuracy: 0.9106 - val_loss: 0.2123 - val_accuracy: 0.9162\n",
            "Epoch 9/10\n",
            "301/301 [==============================] - 20s 66ms/step - loss: 0.2450 - accuracy: 0.9136 - val_loss: 0.2335 - val_accuracy: 0.9204\n",
            "Epoch 10/10\n",
            "301/301 [==============================] - 20s 67ms/step - loss: 0.2450 - accuracy: 0.9119 - val_loss: 0.0951 - val_accuracy: 0.9124\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Qr05khgtg7Wj",
        "colab_type": "code",
        "colab": {}
      },
      "source": [
        ""
      ],
      "execution_count": 0,
      "outputs": []
    }
  ]
}